Wealth taxes are the talk of the town. A way too long "accounting" thread on the French wealth tax, the revenue it raised (EUR5.2bn, 0.24% of GDP in 2014), its limitations, and the lessons it potentially holds for similar plans in the US (1/n).
Data from 2014: last year of the best available estimates of wealth distribution - Garbinti, Goupille-Lebret, and Piketty (2017). Also interesting experiment: K inc taxes then at same level as L inc taxes - "fiscal revolution" by then Socialist president F. Hollande. (2/n)
First, the tax rate schedule: ISF ("impot sur la fortune") was indeed "progressive wealth taxation". 0.5% on the fraction of income from 0.8M to 1.3M; 0.7% from 1.3M to 2.6M; 1% from 2.6M to 5M; 1.25% from 5M to 10M; and 1.5% above 10M. (3/n)
Second, the tax base: in 2014, around 340,000 tax units, c. top 1% of wealth-holding hh (+ on this later). Based on GGP2017, top 1% held c.EUR2,400 billion (23% share of total wealth), threshold of 1.9M (for all assets, including tax-exempt) and average wealth of 4.6M (4/n)
Third, the exemptions: tax on (net-of-debt) household wealth (including unmarried couples to avoid strategic divorce), but exempted art and collectibles, 30% of owner-occupied housing, some - not all - agricultural land (5/n)
Most importantly, the tax excluded: “business assets” (where taxpayer or his/her spouse held controlling interest >25% and played managing role), small business holdings (anything <250 employees or 50M in sales), and some retirement assets (6/n)
GGP2017 estimate share of housing c.15% in 2014, quickly declining within top bracket, and “business assets” at 5 to 10% of wealth in the top bracket. Theoretical upper bound for tax-exempt assets is 15% of the net worth of the top 1% (7/n)
But that’s only in theory, and as we shall see, in practice it seems way more was exempted. Indeed, name of the game for tax advisors: making as much as possible of other assets fall under the “business asset” tax exempt category… (8/n)
Rate schedule and average wealth in top 1% yield lower bound for effective average tax rate at 0.7%; combined with theoretical 85% exemption, yields theoretical lower bound for revenue of c.EUR14bn, in lieu of actual 5.2bn raised. So WTAF happened, and what do we learn? (9/n)
1.Exemptions matter. Not only do they directly exclude taxable wealth, but they also foster incentives to allocate wealth into untaxed assets. In France, anecdotal evidence it supported the housing boom given primary residence partial exemption. (10/n)
2.Capitalization-based measures of top wealth might be overestimates. In France, like in the US, higher returns for the wealthiest on fixed-income assets (and zero tax reporting at the bottom) may imply top 1% hold a smaller share of wealth than estimated (11/n)
3. Fiscal externalities and interactions with other taxes matter. Part of the low revenue comes from a ceiling of 75% average tax on income, forcing the gvt to rebate part of the wealth tax to highest wealth/income ratio hh, especially given high K inc taxes. (12/n)
4. A tax targeting "the rich" might end up not exactly hitting those it initially intended. Does not really matter if exemption threshold is very high, like in the Warren plan, but social welfare weight on people w. "only" 1.5M in wealth need not be exactly 0... (13/n)
Under-reporting/asset allocation/mobility/tax-dodging/consumption-savings choices/charitable giving... True extent of responses is mostly unknown, especially for a large country like the US, where enforcement likely better, but cost of getting it wrong may be larger... (14/14)
